From 71300113feade013c5a22264483c02b69c62be61 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fran=C3=A7ois=20Poulain?= <fpoulain@metrodore.fr>
Date: Tue, 4 Aug 2020 11:22:19 +0200
Subject: [PATCH] =?UTF-8?q?feat(convert):=20am=C3=A9lioration=20des=20nett?=
 =?UTF-8?q?oyages=20de=20logo?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 drupal2spip_lal/base/convert.py | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/drupal2spip_lal/base/convert.py b/drupal2spip_lal/base/convert.py
index d3e6e8a..e02d420 100644
--- a/drupal2spip_lal/base/convert.py
+++ b/drupal2spip_lal/base/convert.py
@@ -75,9 +75,10 @@ def fetch_and_remove_logo(article, force_download):
             for elem in list(img.previous_siblings) + list(img.next_siblings)
             if elem != '\n'
         ]
-        if img.parent.name == 'a' and not has_siblings:
-            img = img.parent
-        img.replace_with('')
+        if img.parent.name in ['a', 'p'] and not has_siblings:
+            img.parent.replace_with('')
+        else:
+            img.replace_with('')
 
     soup = bs(article.descriptif, 'html.parser')
     img = soup.find('img')
@@ -91,6 +92,15 @@ def fetch_and_remove_logo(article, force_download):
         fetch_logo(src)
         remove_img(img)
         article.descriptif = str(soup)
+
+        # L'image est généralement reprise dans le corps avec un format
+        # différent (par ex sans lien vers l'article).
+        soup = bs(article.texte, 'html.parser')
+        img = soup.find('img', src=src)
+        if img:
+            remove_img(img)
+        article.texte = str(soup)
+
         article.save()